"Fossies" - the Fresh Open Source Software Archive  

Source code changes of the file "addons/SVAR/SVAR_compatibility_funcs.inp" between
gretl-2020a.tar.xz and gretl-2020b.tar.xz

About: gretl (Gnu Regression, Econometrics and Time-series Library) is a cross-platform statistical package mainly for econometric analysis.

SVAR_compatibility_funcs.inp  (gretl-2020a.tar.xz):SVAR_compatibility_funcs.inp  (gretl-2020b.tar.xz)
# Backward compatibility functions which eventually should become # Backward compatibility functions which eventually should become
# obsolete # obsolete
# (either because they are in extra.zip, or other reasons) # (either because they are in extra.zip, or other reasons)
##################################
# getstrings_byindex: gretl 2020b will have the built-in feature
# to slice / fancy-index into an array with a vector
function strings getstrings_byindex(const strings S, const matrix indices)
# collects the (in general non-contiguous) sub-array
# indexed by indices (imitate fancy indexing)
strings out = null
matrix m = vec(indices)
loop i = 1..nelem(m) -q
out += S[m[i]]
endloop
return out
end function
### drawnormwis() was called BayesianDraw ### drawnormwis() was called BayesianDraw
### It will go into extra 0.7 in some incarnation. ### It will go into extra 0.7 in some incarnation.
function matrices drawnormwis(const matrix iXX, const matrix B, function matrices drawnormwis(const matrix iXX, const matrix B,
const matrix Sigma, int T[1::]) const matrix Sigma, int T[1::])
# Draw from a standard Normal-(inverse)-Wishart prior # Draw from a standard Normal-(inverse)-Wishart prior
# for a multi-equation regression model. # for a multi-equation regression model.
# #
# iXX: K x K matrix (X'X)^{-1} (same for all equations!) # iXX: K x K matrix (X'X)^{-1} (same for all equations!)
 End of changes. 1 change blocks. 
0 lines changed or deleted 15 lines changed or added

Home  |  About  |  Features  |  All  |  Newest  |  Dox  |  Diffs  |  RSS Feeds  |  Screenshots  |  Comments  |  Imprint  |  Privacy  |  HTTP(S)